Time lost on every screenshot period

[kolorafa]

It's not huge, but user is losing 1s on every screenshot period (sometimes 2s).

So you work from 8:00 to 16:00, have a screenshot every 1min, and when you stop at 16:00 cattr would show on dashboard 7:52 work time.

[kolorafa]

It might be that app is correct, but the dashboard that group the tasks is wrong:

We are fetching the /api/time-intervals/dashboard
and comparing "duration" to time difference from start to stop for the coutned group

The screenshot period is 2minutes, and the code is running in loop with sleep 60;

[2021-10-15 12:53:27] local.INFO: Duration difference for entry id 1180: 16s  
[2021-10-15 12:55:37] local.INFO: Duration difference for entry id 1180: 17s  
[2021-10-15 12:57:48] local.INFO: Duration difference for entry id 1180: 18s  
[2021-10-15 12:58:53] local.INFO: Duration difference for entry id 1180: 19s  
[2021-10-15 13:00:57] local.INFO: Duration difference for entry id 1180: 20s  
[2021-10-15 13:03:40] local.INFO: Duration difference for entry id 1180: 21s  
[2021-10-15 13:05:50] local.INFO: Duration difference for entry id 1180: 22s  
[2021-10-15 13:06:57] local.INFO: Duration difference for entry id 1180: 23s  
[2021-10-15 13:09:07] local.INFO: Duration difference for entry id 1180: 24s  
[2021-10-15 13:11:16] local.INFO: Duration difference for entry id 1180: 25s  
[2021-10-15 13:13:27] local.INFO: Duration difference for entry id 1180: 26s  
[2021-10-15 13:15:36] local.INFO: Duration difference for entry id 1180: 27s  
[2021-10-15 13:17:44] local.INFO: Duration difference for entry id 1180: 28s  
[2021-10-15 13:19:55] local.INFO: Duration difference for entry id 1180: 29s  
[2021-10-15 13:21:00] local.INFO: Duration difference for entry id 1180: 30s  
[2021-10-15 13:23:11] local.INFO: Duration difference for entry id 1180: 31s  
[2021-10-15 13:25:21] local.INFO: Duration difference for entry id 1180: 32s  
[2021-10-15 13:27:32] local.INFO: Duration difference for entry id 1180: 33s  
[2021-10-15 13:29:44] local.INFO: Duration difference for entry id 1180: 34s  
[2021-10-15 13:31:55] local.INFO: Duration difference for entry id 1180: 35s  
[2021-10-15 13:33:01] local.INFO: Duration difference for entry id 1180: 36s  
[2021-10-15 13:35:13] local.INFO: Duration difference for entry id 1180: 37s  
[2021-10-15 13:37:25] local.INFO: Duration difference for entry id 1180: 38s  
[2021-10-15 13:39:37] local.INFO: Duration difference for entry id 1180: 39s  
[2021-10-15 13:41:49] local.INFO: Duration difference for entry id 1180: 40s  
[2021-10-15 13:43:06] local.INFO: Duration difference for entry id 1180: 41s  

[vermorag]

Mostly, it looks like the app issue. As I understand desktop app - it logics looks like that:

1. Finish counting of the current interval
2. Try to send it to server and wait for an answer
3. If server don't answer - backup interval
4. Start recording of the new interval

And since 2 and 3 point waits for server answer it can take up to 2-3 seconds, that causes your issue. I'll try to have a deeper look in code from that side.